Negative resistance distances

[marcabeer]

Hi all,

I have what is hopefully an easy question to answer. I am using CircuitScape to calculate resistance distances for all pairs of cells in simulated raster data (with raster cell values ranging from 1-100). Approximately 50% of the resistance distances in the matrix output by CircuitScape are negative values, which has thrown me for a loop. My understanding is that all resistance distances should be positive, as the value -1 is reserved for infinite resistance (per the user guide). Given that 50% of the values are negative (indeed, the distribution of resistance value is roughly symmetrical around zero), I suspect that some sort of rescaling has happened. However, I haven't been able to find any information regarding this. Can someone please confirm? Perhaps something is astray in my .ini file - I would be happy to provide materials if the answer for this is not simple.

Thank you,
Marc

[ranjanan]

Can you provide some files please? I believe they're supposed to be positive too. Voltages can be positive or negative. Maybe there's a bug where a voltage file is being written into resistances.

[marcabeer]

files_MAB.zip

Thank you for the quick reply! Attached is a folder containing my .ini file, nodes file, and raster surface. I've also included the first 100x100 portion of the resistance distance matrix that I obtain from CircuitScape (rdist.txt; tab-delimited).

Edit to say that this usually takes around 3 minutes to run with 8 cores.